passenger experience

Passenger experience enhancement is a widely discussed topic in aviation. With the radical evolution of the aviation industry in the recent years, passenger experience is not only about the comfort of the passenger during the flight, but it extends to a wider spectrum of facets, not only involving the airline as one would think, but it becomes a more holistic and inclusive subject.

Services for Improved Passenger Experience

Nowadays several parties are involved in the overall passenger experience, from airlines and airports to ground handling companies and caterers, looking to optimize their services and product offerings to achieve this. Cellock provides services to airlines and caterers and there is great interest from their side to evaluate their status and improve any weaknesses spotted; BoB (Buy-on-Board) is a complete inventory and warehouse management platform, used by Airlines and Caterers, to handle all airplane loadings, stock management and supply tracking. BoB is accompanied by the Crew app, facilitating the needs of the flight attendants during the flight. Using the Crew app, the crew can monitor the loaded trays and products stock on-board, report for missing or damaged items as well as receive payments for sales in cash or by cards, and in multi-currencies. Both the system and the Crew app are PCI certified and all data gathered are anonymized, while sensitive data are well secured and protected from unauthorized access.

The Cellock Demonstrator Scenarios  in ICARUS

Cellock as a Demonstrator in the ICARUS Project had two objectives, to be investigated in two different scenarios. Both objectives aimed in enhancing Cellock’s Buy-On-Board (BoB) functionality, using the ICARUS platform. The first scenario was the “Prediction of sales on-board and tray loading suggestions” and the second objective was to “Predict profitable product discounts and offers to increase in-flight sales”.

Scenario #1:  Prediction of sales on-board and tray loading suggestion

In order to develop the first scenario, Cellock extracted insights from historical data taking into consideration different aspects for in-flight sales, in order to more accurately predict the sales per product and product category for each flight. Historical data from BoB on retail and FnB in-flight sales, number of passengers, airplane loading for FnB, flights discrepancies, as well as other related external open data, such as weather data, flight status data, were used at the initial step. In the subsequent stage, the extracted knowledge and insights from the statistical analysis, as well as the aforementioned datasets were combined in order to train a machine learning algorithm for predicting the number of in-flight product sales, and derive to ways of optimizing tray loadings, and minimizing in-cabin waste.

CELLOCK Demonstrator Scenario 1 – Interactions with Aviation Stakeholders

By using the latest version of the ICARUS platform, Cellock executed the necessary applicable test cases in a successful manner and produced usable results.

Scenario #2: Predict profitable product discounts and offers to increase in-flight sale

The second scenario aimed to extend the BoB’s functionality by enhancing its currently basic analytics features. In this demo scenario, the airlines and catering service providers were able to receive suggestions and predictions on product discounts and bundle product offers, in order to increase in-flight sales. Currently, BoB keeps track of the available products and sales data for both the FnB (Food and Beverage) and Duty-free carts on-board.

According to the initial execution planning, the two scenarios were executed consecutively, with the focus of the early and intermediate demonstrator releases on the 1st scenario, while the 2nd scenario ran during the final demonstrator release.

During the Intermediary demonstrator, the implementation procedure followed the steps below:

  1. Created Virtual Datasets with own data from BoB
  2. Purchased Data from OAG: connected to Ethereum wallet, negotiated the contract, and bought the dataset.
  3. Linked own data with purchased data assets
  4. Designed and applied descriptive and predictive analytics: The analytics workflow was defined by setting the analytics methods and the required datasets upon applying the necessary data manipulation functions.
  5. Results and visualization: The results were displayed on the ICARUS platform and could also be visualized by offering a quick review to the user.

Key Take-aways

In conclusion, the execution of the two scenarios in hand, was conducted successfully, and the involved stakeholders now have a more solid and data-backed method to “Predict sales on-board and optimize tray loading” and “Predict profitable product discounts and offers to increase in-flight sales”. The pain point that existed in this particular operation in the aviation industry, has now been bridged and the results, not only benefit the airlines in terms of securing higher ancillary revenues and savings, but at the same time may enhance the overall passenger experience.

Blog post prepared by CELLOCK

Featured Photo by Ethan Hu on Unsplash